AN ADAPTIVE AND SELF-COGNITIVE ACE FILTER

Abstract

In Automatic Generation Control systems, the Area Control Error, (ACE) is a parameter which is used as a measure of the generation requirement of an area in a multi����area inter-connected power system. Due to normal load changes, ACE signals are random, in nature and, contain wide band noise. Filtering of ACE signals to reduce the unwanted noise components is needed to reduce mechanical wear and provide smoother operation. The classical ACE filters now in use ate incapable of achieving a high noise, reduction ratio together with a fast transient response. In this thesis, the application of Kalman filtering of ACE signals is discussed and an adaptive ACE filter is described which achieves both a high noise reduction ratio and fast response with excellent flexibility for varying noise conditions. The transient performance has been enhanced by applying a pattern recognition, technique. A design scheme is provided for practical applications, and some implementation problems are assessed.